• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

FEATURED Marathon SIII: December SuperPi 32M

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
sub 6 min or it doesn't count

:) I was hoping to see high 5 min on ivy this country cup but I doubt I see that

I benched the 3570k from my HTPC did not go so good could not break 2300 with the micron d9's booted at 2600 with my bad 3570k and couldn't tighten up the 2400 sam's at all -should get my new set's tomorrow then I should know then if it's just a weak IMC

here's the best I thought the micron d9"s could do http://hwbot.org/submission/2340872_quickfast_superpi___32m_core_i5_3570k_6min_36sec_609ms that is before I was on the team and had no clue about overclocking ram so I was hoping to shave 20 seconds off the time most likely wishful thinking, but I love those suicide runs :)
 
I'm trying to make something more on 5820k but high latency of 4 channels is not helping. I see that 1150 haswells are better in Spi. Heh, I just noticed that my best spi32 result on hwbot is on pentium :p

Older microns are bad on air. Some better batches are good for LN2 but BBSE/Hypers/PSC are still better and easier to OC. New D9 are generally not making more than ~2200 but can push them to make 8-8-8 or something near. You had some special kit if you made it run at 2600. I actually saw couple of these kits running at 2600 9-9-9 and high voltages.

edit:
improved my result about 15 secs and trying some more :)
 
Last edited:
Delete this post please.
 

Attachments

  • Capture.PNG
    Capture.PNG
    475.2 KB · Views: 82
  • Capture2.PNG
    Capture2.PNG
    559.4 KB · Views: 83
Last edited:
here is one of mine. only on air so not going to be very fast. but not horrible iether supi4900-6-33-996.jpg
 
And there it is! :D

New 5820K, 5GHz on water ~1.425V vs old which could make 4.8GHz ~1.6V. That's on 2 cores. For all 6 it needs some more volts but I had no time to check it above 4.7GHz. XTU is passing @4.7GHz 1.325V on all cores+HT.
Also cache runs 200MHz higher at lower voltage.
Now I hope it won't die in next 2-3 weeks like all other good chips I had.

@peanutbudder
This kit will probably make something like 2933 11-14-14 or 12-14-14 at reasonable voltages. With some luck you can see 3000 12-14-14 1.65-1.75V. It's also double sided hynix so pefromance will be good regardless of clock.
Can also try 2400 10-12-12 1.65V or 2600 10-12-12 1.65-1.75V. My better Patriot kit at similar IC could make something like this.
 
Nice run woomack, guess that's where the quad channel helps a bit.

@ dejo, I updated the results but could you please use the format I asked for when submitting.
 
Nice run woomack, guess that's where the quad channel helps a bit.

Somehow I can't see that quad channel in spi32 :p ... higher bandwidth but lower access time. 1150 haswells are about the same, even though memory bandwidth is about 50% worse.
 
New 5820K, 5GHz on water ~1.425V vs old which could make 4.8GHz ~1.6V. That's on 2 cores. For all 6 it needs some more volts but I had no time to check it above 4.7GHz. XTU is passing @4.7GHz 1.325V on all cores+HT.
Also cache runs 200MHz higher at lower voltage.
Now I hope it won't die in next 2-3 weeks like all other good chips I had.
Looks like a good chip Woomack, fingers crossed for ya.
 
Back